The deployment of deep convolutional neural networks (CNNs) in SAR ship real-time detection is largely hindered by huge computational cost. In this paper, we propose a novel learning scheme for training a lightweight ship detector called Tiny YOLO-Lite, which simultaneously 1) reduces the model storage size; 2) decreases the floating point operations (FLOPs) calculation; 3) guarantees the high accuracy with faster speed. This is achieved by self-designed backbone structure and network pruning, which enforces channel-level sparsity in the backbone network and yields a compact model. In addition, knowledge distillation is also applied to make up for the performance decline caused by network pruning. Hereinto, we propose to let small student model mimic cumbersome teacher's output to achieve improved generalization. Rather than applying vanilla full feature imitation, we redefine the distilled knowledge as the inter-relationship between different levels of feature maps and then transfer it from the large network to a smaller one. On account that the detectors should focus more on the salient regions containing ships while background interference is overwhelming, a novel attention mechanism is attached to the distilled feature for enhanced representation. Finally, extensive experiments are conducted on SSDD, HRSID and two large-scene SAR images to verify the effectiveness ofthe thinner SAR ship object detector in comparison of with other CNN-based algorithms. The detection results demonstrate that the proposed detector can achieve lighter architecture with 2.8M model size, more efficient inference (>200fps) with low computation cost, and more accurate prediction with knowledge transfer strategy.

Simple linear iterative clustering (SLIC) algorithm was proposed for superpixel generation on optical images and showed promising performance. Several studies have been proposed to modify SLIC to make it applicable for polarimetric synthetic aperture radar (PolSAR) images, where the Wishart distance is adopted as the similarity measure. However, the superpixel segmentation results of these methods were not satisfactory in heterogeneous urban areas. Further, it is difficult to determine the tradeoff factor which controls the relative weight between polarimetric similarity and spatial proximity. In this research, an adaptive polarimetric SLIC (Pol-ASLIC) superpixel generation method is proposed to overcome these limitations. First, the spherically invariant random vector (SIRV) product model is adopted to estimate the normalized covariance matrix and texture for each pixel. A new edge detector is then utilized to extract PolSAR image edges for the initialization of central seeds. In the local iterative clustering, multiple cues including polarimetric, texture, and spatial information are considered to define the similarity measure. Moreover, a polarimetric homogeneity measurement is used to automatically determine the tradeoff factor, which can vary from homogeneous areas to heterogeneous areas. Finally, the SLIC superpixel generation scheme is applied to the airborne Experimental SAR and PiSAR L-band PolSAR data to demonstrate the effectiveness of this proposed superpixel generation approach. This proposed algorithm produces compact superpixels which can well adhere to image boundaries in both natural and urban areas. The detail information in heterogeneous areas can be well preserved.

Ship detection in polarimetric synthetic aperture radar (PolSAR) images has attracted widespread attention in recent years. However, pixel-level detection methods are heavily affected by inherent speckle noise. In this letter, we proposed a detection method that enhances the ship-sea contrast beforehand by combining local statistical saliency and scattering mechanism coherence in superpixel-level. First, simple linear iterative clustering (SLIC)-based segmentation method is adopted for PolSAR images to generate superpixels. Then, local saliency is calculated based on superpixel-level similarity from the perspective of statistical characteristics. Based on this, the superpixel-level modified polarimetric coherence metric is obtained from the perspective of physical scattering mechanisms, which can help distinguish small ships with low saliency and strong sea clutters with high saliency. Ship detection is achieved by combining the two features above. The experimental results based on real PolSAR data show that compared with other classic and state-of-the-art methods, the proposed method has improved the figure of merit (FoM) by at least 4.28% and has increased the target clutter ratio by at least 8.43 decibel (dB) on average.

Synthetic aperture radar (SAR) image target detection methods based on semi-supervised learning, such as the mean teacher framework, have shown promise in diminishing the issue of limited labeled data. However, several challenges exist in current methods. Firstly, data augmentation techniques designed for optical images may not be suitable for SAR images due to differences in imaging methods. Additionally, the contribution of pseudo labels remains constant during the initial retraining stage can lead to degradation in prediction results. Moreover, the low quality of predicted bounding boxes poses a challenge for effective retraining. To address these challenges, we propose an end-to-end semi-supervised detection method based on the mean teacher framework. To enhance the robustness of training, we firstly introduce SAR-specific data augmentation techniques, including multiplicative noise, which effectively increase the diversity of training samples. Secondly, we propose a method that weights the losses of pseudo-labeled data using a hard-sigmoid function, gradually improving the importance of pseudo-labeled data during retraining, thereby alleviating their potential negative impact on the training process. Finally, we propose an IoU-aware subnetwork to incorporate high-quality pseudo-labeled bounding boxes into retraining, allowing them to contribute to network adjustments while mitigating the impact of low-quality samples. Experimental evaluations on publicly available SAR image datasets demonstrate the effectiveness of our proposed method in improving the target detection capability of semi-supervised SAR target detection.

Aircraft detection in synthetic aperture radar (SAR) images is one challenging task due to the discreteness of aircraft scattering, the diversity of aircraft size, and the interference of background. In order to deal with these problems, a novel method named scattering enhancement and feature fusion network (SEFFNet) is here proposed to detect aircraft via combining traditional image processing and deep learning together. At first, a scattering information extraction and enhancement module (SIEEM) is proposed to highlight the scattering points of aircraft targets. Then, to more effectively focus on the location of aircraft targets, a space-to-depth coordinate attention module (SDCAM) is further designed, following which an efficient multi-scale feature fusion pyramid (FFP) is also introduced to fuse the semantic information of different layers. At last, a contextual fusion head (CFH) is built to improve the receptive field for better detecting aircraft. The experiments carried out on the popular datasets SADD and SAR-AIRcraft-1.0 show that SEFFNet is more appropriate for aircraft detection, especially the small-size aircraft detection, in comparison with other state-of-the-art (SOTA) methods. Taking the dataset SADD for example, on average, the precision, recall, F1-score, and APs values are respectively 2.8%, 2.6%, 2.7%, and 2.0% higher than the baseline network YOLOv5.

Deep learning methods have made significant progress in the field of synthetic aperture radar (SAR) target recognition, effectively extracting target image domain features. However, they generally overlook the inherent electromagnetic scattering (ES) characteristics of targets, which limits recognition accuracy and physical interpretability. To address this issue, this paper proposes a multi-azimuth scattering feature fusion network (MASFF-Net) that integrates ES features with image domain features for SAR target recognition. Initially, we design a multi-azimuth scattering feature extraction module (MASF) to acquire SAR sub-aperture image data from multiple azimuths and capture the strong scattering point (SSP) features of multi-azimuth SAR images, whose spatial distribution characterizes the geometric structure information of the target. Subsequently, we propose a hierarchical multi-scale feature extraction module (HMFE) to explore the global semantic and local detail features of the target. Finally, we design a scattering-guided feature fusion module (SGFF), which achieves the organic fusion of ES features and image features under the guidance of geometric scattering prior information, and enhances key region features using a multi-branch coordinate attention mechanism. Experiments on a self-constructed measured dataset of 10 types of SAR vehicle targets, the moving and stationary target acquisition and recognition (MSTAR) benchmark dataset, and the SAR-AIRcraft-1.0 aircraft dataset demonstrate that this method achieves higher recognition performance while maintaining physical interpretability.